Physiological and morphological responses of ... - CSIRO Publishing

CSIRO PUBLISHING

Functional Plant Biology, 2004, 31, 181–194

www.publish.csiro.au/journals/fpb

Physiological and morphological responses of grassland species to elevated atmospheric CO2 concentrations in FACE-systems and natural CO2 springs Susanna MarchiA,D, Roberto TognettiB, Francesco Primo VaccariA, Mario LaniniA, Mitja Kaligarič C, Francesco MigliettaA and Antonio RaschiA AIstituto

di Biometeorologia, Consiglio Nazionale delle Ricerche, I-50144 Firenze, Italy. di Scienze Animali, Vegetali e dell’Ambiente, Università del Molise, I-86100 Campobasso, Italy. CDepartment of Biology, Pedagogical Faculty, University of Maribor, SI-2000 Maribor, Slovenia. DCurrent address: Scuola Superiore Sant’Anna di Studi Universitari e Perfezionamento, I-56100 Pisa, Italy. Corresponding author; email: [email protected]

BDipartimento

Abstract. Stomatal density, leaf conductance and water relations can be affected by an increase in the concentration of atmospheric CO2, and thus affect plant productivity. However, there is uncertainty about the effects of elevated CO2 on stomatal behaviour, water relations and plant productivity, owing to the lack of long-term experiments in representative natural ecosystems. In this work, variations in stomatal density and index, leaf water relations and plant biomass of semi-natural grassland communities were analysed under field conditions by comparing plants in three different experimental set-ups (natural CO2 springs, plastic tunnels and mini-FACE systems). Natural degassing vents continuously expose the surrounding vegetation to truly long-term elevated CO2 and can complement short-term manipulative experiments. Elevated CO2 concentration effects on stomata persist in the long term, though different species growing in the same environment show species-specific responses. The general decrease in stomatal conductance after exposure to elevated CO2 was not associated with clear changes in stomatal number on leaf surfaces. The hypothesis of long-term adaptive modifications to stomatal number and distribution of plants exposed to elevated CO2 was not supported by these experiments on grassland communities. Elastic cell wall properties were affected to some extent by elevated CO2. Above-ground biomass did not vary between CO2 treatments, leaf area index did not compensate for reduced stomatal conductance, and the root system had potentially greater soil exploration capacity. Considerable between-species variation in response to elevated CO2 may provide a mechanism for changing competitive interactions among plant species. Keywords: biomass, leaf conductance, root density, stomatal density, water relations. Introduction The steady increase in atmospheric CO2 concentration is well documented (Keeling et al. 1995). Stomates are integrators of all environmental factors affecting plant growth (Morison 1998), and changes in stomatal behaviour may contribute to altering ecosystem water use, carbon gain and yield in natural grasslands (Jackson et al. 1994). Generally, stomatal conductance is expected to decline in herbaceous plants with a rise in CO2 above current levels (Field et al. 1995; Ward et al. 1999). Initial observations suggested that a reduction in stomatal density with increasing CO2 concentration is a general response by plants (Woodward 1987). However, a wide range of specific stomatal responses to changes in CO2 concentration has

been reported (Amthor 1995). Indeed, experiments have shown declines, no consistent effects and even increases in stomatal density as CO2 concentration increases (Ferris and Taylor 1994; Knapp et al. 1994; Woodward and Kelly 1995). This wide variability of responses makes the identification of mechanisms difficult because of different interspecific genetic backgrounds (Woodward et al. 2002). A major factor that could alter the response of stomata to CO2 concentration is the degree to which they acclimate functionally and morphologically. This would be ecologically important if it either tempered or enhanced the reduction in stomatal conductance to water vapour with rising CO2 concentration (Morison 1998). Acclimation and adjustment could have significant impacts on gas exchange and water

Abbreviations used: FACE, free air CO2 enrichment; gs , stomatal conductance; PVC, pressure–volume curve. © CSIRO 2004

10.1071/FP03140

1445-4408/04/020181

182

Functional Plant Biology

relations, and thus, on plant productivity (Seneweera et al. 2002; Hovenden 2003). Commonly, plant growth is stimulated by an increase in CO2 concentration, although mixtures of species may either have similar gains in productivity (Overdieck and Reining 1986) or different levels of stimulation (Curtis et al. 1989). Natural grasslands cover approximately 24 × 106 km2 of the earth’s surface (Parton et al. 1993), accounting for roughly 12% of the total carbon storage in the continental biosphere (Schlesinger 1991). Low-input pastoral agriculture in temperate regions is based on grassland communities, and there is practical interest in any environmental change, such as an increase in atmospheric CO2 concentration, that might alter their stability. Current predictions about the effect of rising CO2 concentration on stomatal responses of grassland communities are often based on data from experiments performed under controlled conditions (growth chambers, glasshouses, open-top chambers and field enclosures). These experiments suffer from poor atmospheric coupling and an altered balance between energy supply and water loss from leaves (Knapp et al. 1994). Observations made with herbaria specimens, fossil vegetation and archaeological material might be confounded by consequences of environmental variables other than increased CO2 concentration (Körner 1988). Questions of acclimation and adaptation to elevated CO2 suggest a need for measurements made under the realistic conditions of Free Air CO2 Enrichment (FACE) experiments, conceived to exclude environmental disturbances between entire plant communities exposed to high CO2 and external control plants (Hendrey et al. 1993). Although not ideal experiments, natural geological CO2 springs have, in some cases, exposed the terrestrial ecosystem to high CO2 for many years (Miglietta et al. 1993; Newton et al. 1996; Cook et al. 1998). Advantages of natural CO2 springs include the absence of alteration to plant micro-environment and artificial soils, a large plot area exposed to elevated CO2 facilitating ecosystem-level studies, the occurrence of natural climatic cycles and biotic interactions, and the long history of high CO2 in an otherwise natural environment. Drawbacks of these springs are the potential high fluctuations in CO2 concentration on space and time scales, the uncertainty of true replicates and controls, spatial variability in the soil environment and, often, the presence of gaseous pollutants whose effects may be considered as experimental artefacts. Long-term experimental manipulations of CO2 level designed to address acclimation responses are technically demanding and expensive to operate. An alternative is to use, in parallel with controlled FACE experiments, naturally occurring high CO2 treatments as substitutes for long-term elevated CO2 trials. It remains uncertain whether or not a decrease in leaf conductance under elevated CO2 will be accompanied by changes in stomatal number. Points to be elucidated include differences between short-term and long-term effects of

S. Marchi et al.

increased CO2 on stomata, acclimation of stomata to elevated CO2 (Drake et al. 1997), integration of information on stomatal behaviour and growth performance, variation in the response to increased CO2 of stomata on the two different leaf surfaces (Pearson et al. 1995). Morphogenetic constraints on leaf growth seem to regulate final leaf area synchronously with environmental factors (Van Volkenburgh 1999), and increases in growth have been associated with higher cell production rates (Beemster and Baskin 1998). In this study we assessed the variability in stomatal response, growth and water relations of semi-natural grassland communities grown under three different experimental conditions. (1) A grassland community grown at a natural geological CO2 spring compared with corresponding plants grown at a nearby control site, (2) plants originated by a natural grassland community acclimated to elevated CO2 in a natural CO2 spring then transplanted into a mini-FACE system facility, including control and CO2 enriched rings, and (3) a grassland community grown in mini-FACE system rings, under ambient or elevated CO2. Materials and methods The natural CO2 spring at Strmec Part of the study was conducted under natural conditions in Slovenia at the CO2 degassing vent at Strmec near Gornja Radgona (46°39′ N, 16°00′ E). The area around Gornja Radgona and Radenci is known for its geothermal activity associated with water springs, due to geothermal anomalies determining a gaseous lift from deep fractures. The climate at the site is temperate continental with high thermal excursion that ranges from 20 to 25°C during the year. Average annual precipitation is 800–1200 mm, principally falling as rain in summer months, while winters are cold and snowy. The total surface of the enriched area is approximately 500 m2. There are many gas emissions, which can be noted as rings with reduced vegetation. Two main gas vents are in the ditch at the side of the Ivanjševci–Stavešinci asphalt road, and they are clearly visible when covered with bubbling water; there are also emission sites in the field to the other side of the road. The gas emitted by the vents is 99% CO2 without any H2S or H2SO4, which makes the site particularly suitable for studying the effects of elevated CO2 on healthy vegetation. Further details are given elsewhere (Vodnik et al. 2002; Maček et al. 2003). Since 1996, CO2 concentrations have been continuously monitored and logged in field surveys within the enriched area with a portable single-beam, self-zeroing infrared gas analyser (EGM-1, PP-Systems, Hitchin, UK). Air was sampled at 1-min logging intervals. Additional measurements were made in the period 1997–1999 with CO 2-sensitive diffusion tubes (Dräger, Lübeck, Germany) of up to 1.5 m in height, placed at 30 cm spacing near plant-sampling positions within the area. These tubes provided an estimate of average CO2 concentration integrated over a maximum of 8 h. Repeated readings can be made by visual estimation of colour changes of a specific reactant along a graded scale and readings are corrected for temperature and atmospheric pressure. Plants around the CO 2 spring are exposed to a daytime CO2 concentration of approximately 700 µmol mol–1 throughout the year with short-term variations between 500 and 1000 µmol mol–1 depending on wind speed, temperature, solar radiation and convective turbulence. In conditions of low turbulence CO2 concentration of the enriched area increases with respect to the background average, while values do not change appreciably when

Effects of elevated CO2 on grassland species

winds are strong. During the night, CO 2 concentration increases along the ditch, which acts as a temporary fence trapping the emitted gas. In these conditions CO2 concentration is very high, being lethal to birds, small mammals and insects, which are often found dead at the bottom of the ditch. The area around the CO2 spring can be divided into eight sub-units based on floristic structure. Close to the spring, at the bottom of the ditch where CO2 accumulates for longer periods in the absence of air turbulence, vegetation is scarce owing to the very high CO2 concentration. The typical Epilobium–Junctum effusi association of the vegetation around the CO2 spring is common to nearby areas with ambient CO2 concentration. Nevertheless, as air CO2 concentration changes, the type of vegetation also varies due to different competitive relationships among species. In August and September 1999, measurements were obtained for individuals growing in close association in proximity to the CO2 spring. Additional measurements were made at a control site chosen 300 m from the spring for the similar soil conditions and sun exposure, but where CO2 concentrations were at ambient level (360 µmol mol–1). Sampling was on plants selected from among those species widely present at both the CO2 enriched and control site: Tanacetum vulgaris L. (Composite), Polygonum hydropiper L. (Polygonaceae), Echinochloa crus-galli L. (Poaceae), Rumex crispus L. (Polygonaceae), Plantago lanceolata L. (Plantaginaceae) and Trifolium pratense L. (Fabaceae). Tunnel assembly in the Bossoleto CO2 spring A manipulative experiment was conducted in the proximity of the Bossoleto CO2 spring close to Rapolano Terme (Siena, Italy, 43°17′ N, 11°35′ E and 350 m above sea level); detailed information on the site is given in Körner and Miglietta (1994). Several CO2 vents occur both at the bottom and on the flanks of a circular crater; CO2 concentration gradients are enhanced under stable (windless) atmospheric conditions. The CO2 concentration above experimental plots during the day ranged from 500 to 1000 µmol mol–1 with rapid fluctuations. The H2S and SO2 concentrations in the spring are very low and not considered harmful to plants. Between 1994 and 1996, small plots of vegetation within the crater were enclosed, in situ, in plastic tunnels and exposed to CO2 concentrations of 360 µmol mol–1 or 700 µmol mol–1. Four tunnels, each of 2.5 m2 surface, were assembled with a plastic frame covered by a polycarbonate film (Propafilm-C, ICI, Milan, Italy). The experiment was set up to assess if the effect of stable elevated CO2, in the absence of noxious pollutants, might differ from that of natural fluctuating enrichment. The two CO2-enriched tunnels were fumigated with pure CO2, from gas cylinders, mixed with ambient air from outside the spring; the two control tunnels were fumigated with the same external ambient air. Temperature and CO2 concentration inside the tunnels were continuously monitored by means of thermistors and infrared gas analysers (EGM-1, PP-System, Hitchin). In order to minimise the greenhouse effect of the plastic covering, a constant flux of air was maintained inside each tunnel to keep the temperature near ambient. Experimental details are described by Vaccari et al. (2001). At the end of April 1996, 32 intact clods (50 cm wide, 50 cm long and 15 cm deep) were removed from plots under the tunnels and implanted outside the CO2 spring in a mini-FACE system (see below for description). The air in each of eight circular rings of 1 m in diameter was sampled at 1-min intervals and CO2 concentration controlled by a computerised system; four rings were maintained at ambient CO2 (360 µmol mol–1) and the other four enriched (620 µmol mol–1) with pure CO2 mixed with ambient air. In June 1996, sampling was done on plants selected among those species widely present within both the tunnels and the mini-FACE system, and in CO2 enriched and control atmospheric conditions: Arabis hirsuta L. (Cruciferae), Hippocrepis comosa L. (Fabaceae), Plantago


183

lanceolata L. (Plantaginaceae), Scabiosa columbaria L. (Dipsacaceae) and Silene vulgaris (Moench.) Garcke (Caryophyllaceae). Mini-FACE system details A FACE experiment was performed at a grassland community on an unmanaged, naturally regenerating old field, near Rapolano Terme. The site was uniform in terms of sun exposure and weathering. The soil had not been fertilised for at least 5 years. Additional information can be found in Sanità di Toppi et al. (2002). Plants grew in mini-FACE rings (1.8 m in diameter, toroidal distribution), under ambient (360 µmol mol–1) or elevated CO2 (approximately 580 µmol mol–1). Six enriched and six identical control rings were positioned 5 m apart from one another. Fumigation started in October 1998. The vegetation on the experimental field was harvested simultaneously in all rings in 1999 (the experiment continued in 2000). In summer, each ring (regardless of treatment) was irrigated daily with the same amount of water. Miglietta et al. (1997, 2001) have described previously the mini-FACE technology and its performance. Purified and concentrated CO2 were supplied directly by Messer Italia (Rapolano Terme, Italy), through an underground pipe system. Sampling was done in June, July and August 1999 (depending on the species), when plants were fully developed, on those species widely present in both CO2 enriched and control rings: Avena barbata Pott ex Link (Poaceae), Cynodon dactylon L. (Poaceae), Convolvolus arvensis L. (Convolvulaceae), Trifolium repens L. (Fabaceae), Medicago arabica L. (Fabaceae) and Potentilla reptans L. (Rosaceae). Cynodon dactylon, a C4 weed of Mediterranean environments, was particularly widespread during summer months. Selected plants approximated each other as closely as possible with respect to dimensions and health conditions. Further measurements were taken in 2000 for ecophysiological parameters. Plant measurements Stomatal counts were made on at least two sunny, fully expanded and healthy leaves per plant, which were selected according to uniformity of appearance, growth habit and sun exposure. For each species 12 individuals were sampled in the CO2 enriched environments and 12 in the corresponding controls. Leaf impressions were taken from disks (1.5 cm diameter) cut in the central part of the leaf surface, avoiding major veins. Good quality leaf replicas were obtained using an acetone-soluble transparent plastic material (Rodoith, Weber Metaux, Paris, France) as described by Vazzana et al. (1988). Leaf impressions were taken on abaxial and adaxial leaf surfaces, depending on the morphological characteristics of the species. The number of stomata was counted on microphotographs under three fields of view: 1.3824 mm2 at 100×, 0.3456 mm2 at 200× and 0.0864 mm2 at 400× magnification. For all species leaf impressions were of sufficient quality to distinguish epidermal cells. This enabled calculation of the stomatal index (SI) as: SI = [(no. of stomata) /(no. of stomata + no. of epidermal cells)] × 100. Stomatal index (indicating whether stomatal density has been altered independently of epidermal cell density rather than passively as a result of effects on cell expansion) was not determined for Tanacetum vulgaris, Silene vulgaris, Arabis irsuta and Scabiosa columbaria because of the extremely dense hair layer, which was impossible to remove without damaging the leaf. Leaf conductance (gs) was measured on several leaves per individual of the same species sampled for stomatal counts, with a null-balance steady-state porometer (Li-1600, Li-Cor Inc., Lincoln, NE). Plants growing in CO2-enriched and control environments were sampled at different times during sunny days in summer, when photosynthetic active radiation was greater than 1000 µmol m–2 s–1. The measurement periods differed for atmospheric conditions (average temperature and relative humidity: 24°C and 60% in June, 32°C and

184


S. Marchi et al.

35% in July), less for soil water availability. Data from diurnal time courses indicated that maximum daily light-saturated gs generally occurred in the early morning. Vapour pressure deficit was estimated by measuring relative humidity and temperature with the porometer chamber held open next to the foliage.

In the mini-FACE system, during July 2000, diurnal trends in leaf water potential (pressure chamber) and conductance (Li-1600 porometer) were followed for Cynodon dactylon and Trifolium repens. The pressure–volume approach was used to characterise tissue water relations of Cynodon dactylon and Medicago arabica during July 1999

A

7000 6000 5000

20

4000 3000 2000 1000

15

0

5

10

15

20

25

30

2000

B

1800 1600 1400

20

1200 1000 800 600

15

400

5

10

15

20

25

30

4000

C

3600 3200 2800

20

2400 2000 1600 1200 800

15

400

5

10

15

20

25

30

Fig. 1. Horizontal distribution of day time CO2 concentration in the ditch at the CO 2 spring of Strmec, measured in April 1998 (A), and September 1998 (B) and 1999 (C). Both y- and x-axis are marked in m; different shading in the legend indicates gradients in CO2 concentration (µmol mol–1).



A

20

15

7000

10

6000 5000 4000

5

3000 2000 1000 0

0 0

5

10

15

20

25

30

24 B 22 10 16

20 18

X 1000 20 18 16 14 12 10 8 6 4 2 0

16 2

4

6

8

10

12

14

16

18

22

24

C

200

Height (cm)

20

150 100 50 0 0

500

1000

1500

2000

2500

CO2 concentration (µmol mol–1) Fig. 2. Horizontal distribution of day time CO2 concentration in the ditch and in the field (A, September 1999) and night time CO2 concentration in the field (B, September 1999) at the CO2 spring of Strmec. Both y- and x-axis are marked in m; different shading in the legend indicate gradients in CO2 concentrations (µmol mol–1). Vertical profile of day time CO2 concentration in the middle of the vent (C, September 1999).

185

186


S. Marchi et al.

and 2000. Measurements of pressure–volume relationship and data analysis were performed as described by Tognetti et al. (2000). Samples were collected in early morning and re-hydrated in distilled water before the start of each pressure–volume curve for 180 min, preventing complete saturation of tissues and abnormal shifts in cell wall elasticity and artificially high osmotic potentials. Dry mass (Wd) of leaf tissue was measured after oven drying at 70°C to constant mass. The total water volume lost from the sample at each water potential (Ψ) value was computed by subtracting the fresh mass (Wf) at that point from the saturated mass (assuming a density of 1000 kg m –3). We extrapolated back to compute the fully turgid mass (Ws) of the tissue. Data were plotted as 1/Ψ against 1 – R* (relative water deficit), where R* is the relative water content:

R* =

Wf – Wd . Ws – Wd

(1)

Values of osmotic potential at full hydration (πsat) and at zero turgor (πtlp), R* at zero turgor (R*tlp) and symplastic water fraction (Rs) were calculated following the PVC method described by Schulte and Hinckley (1985). A normal pressure–volume curve consisted of 11–13 data points for each sample. The bulk modulus of elasticity (ε) was calculated from pressure–volume curves. Total aboveground dry mass, as an index of productivity, was measured by weighing plant materials after 48 h in the oven at 70°C; harvested vegetation within the rings, in May, August and October 2000, had previously been divided into Poaceae, Fabaceae and other species. Contemporarily, cumulative leaf area index within rings was estimated through leaf area determinations with a planimeter (Li-3000, Li-Cor Inc.). Root biomass and root density were determined in June and October 2000, at two different soil depths (0–15 cm and 15–30 cm) in each ring. Statistical analysis Data analysis for stomatal density and index was performed in two blocks (adaxial and abaxial surface values) per species. Stomatal

density and index, leaf conductance, water relations and biomass were subjected to one-way analysis of variance; a hierarchical analysis of variance using one fixed factor (CO2 exposure) and one random factor (stomatal density, stomatal index, leaf conductance, water relation parameters or plant biomass for each species) was performed using the SAS statistical package (SAS Inc., Cary, NC).

Results Mean atmospheric CO2 concentrations, measured by diffusion tubes around the CO2 spring in Strmec are represented in Figs 1 and 2. Measurements were conducted on different dates during the growing season (day- and night-time sampling). Horizontal and vertical gradients showed the effective enrichment of the area selected for vegetation sampling. Mean atmospheric CO2 concentration in the miniFACE systems was approximate to the target values, ranging between 560 and 600 µmol mol–1, while it averaged 620 µmol mol–1 for the Bossoleto experiment. At the CO2 spring in Strmec, stomatal density of Tanacetum vulgaris and Rumex crispus (either adaxial or abaxial leaf surface) was not affected by long-term exposure to elevated CO2 (Table 1). Stomatal density of Trifolium pratense resulted as significantly higher in plants grown close to the CO2 spring than at the control site on both leaf surfaces. Stomatal density on the abaxial leaf surface of Plantago lanceolata was significantly higher under elevated CO2; conversely, stomatal density on the adaxial leaf surface of Polygonum hydropiper was significantly lower. Differences in stomatal index between CO2 enriched and control

Table 1. Stomatal density in leaves (either on adaxial or abaxial surface) sampled at the CO2 spring of Strmec (September–August 1999), at the Bossoleto experimental set-up (end of June 1996) and in the mini-FACE system (summer 1999, month depending on the species) Data are the means ± s.e. (number of stomata mm –2). Percentage change between CO2 treatments [(elevated CO2 – ambient CO2)/ambient CO2] and P-level are also reported (CO2 concentrations are reported in the text) Abaxial ambient CO2 elevated CO2 Strmec Tanacetum vulgaris Rumex crispus Plantago lanceolata Polygonum hydropiper Trifolium pratense Bossoleto Hypocrepis comosa Silene vulgaris Arabis irsuta Plantago lanceolata Scabiosa columbaria Mini-FACE Cynodon dactylon Avena barbata Convolvolus arvensis Potentilla reptans Trifolium repens Medicago arabica

∆ (P-level) %

Adaxial elevated CO2 ambient CO2

∆ (P-level) %

95.33 ± 3.63 60.75 ± 3.04 239.31 ± 11.68 173.61 ± 6.09 279.52 ± 9.34

94.00 ± 3.68 54.83 ± 4.81 201.95 ± 11.42 189.72 ± 6.09 234.30 ± 8.92

1.4 (ns) 10.8 (ns) 18.5 (0.027) 9.3 (ns) 19.3 (0.001)

73.58 ± 3.59 41.82 ± 2.51 183.63 ± 8.13 20.28 ± 1.90 271.91 ± 7.47

80.14 ± 3.59 42.00 ± 2.51 189.6 ± 7.63 27.31 ± 1.9 237.56 ± 7.63

–8.2 (ns) –0.4 (ns) –3.1 (ns) –25.7 (0.011) 14.5 (0.002)

132.27 ± 7.3 83.8 ± 4.18 162.8 ± 11. 226.73 ± 12.59 164.37 ± 5.68

114.08 ± 6.02 98.07 ± 6.02 146.07 ± 5.29 210.66 ± 9.83 177.16 ± 7.09

15.9 (ns) –14.6 (ns) 11.5 (ns) 7.6 (ns) –7.2 (ns)

169.83 ± 8.76 92.93 ± 4.08 123.85 ± 6.86 201.18 ± 13.14 116.61 ± 10.03

149.6 ± 8.97 96.81 ± 7.7 85.16 ± 6.71 161.45 ± 5.36 88.17 ± 5.53

13.5 (ns) –4.0 (ns) 45.4 (0.0002) 24.6 (0.0057) 32.2 (0.0145)

194.91 ± 6.89 54.62 ± 2.88 89.23 ± 3.57 444.13 ± 13.93 177.23 ± 7.01 187.73 ± 9.27

210.41 ± 6.99 57.24 ± 3.00 89.94 ± 3.57 408.88 ± 14.12 168.45 ± 6.91 203.39 ± 9.27

–7.4 (ns) –4.6 (ns) –0.8 (ns) 8.6 (ns) 5.2 (ns) –7.7 (ns)

185.26 ± 7.95 62.75 ± 5.52 51.78 ± 3.16 55.55 ± 4.17 376.39 ± 11.93 253.72 ± 9.9

199.30 ± 7.95 63.66 ± 5.21 51.76 ± 3.16 56.02 ± 4.17 460.39 ± 11.93 294.22 ± 10.04

–7.0 (ns) –1.4 (ns) 0.0 (ns) –0.8 (ns) –18.2 (0.0001) –13.8 (0.005)



site resembled to some extent those observed in stomatal density (Table 2). At the Bossoleto experimental set-up, stomatal density did not vary significantly between treatments on both leaf surfaces of Hippocrepis comosa and Silene vulgaris (Table 1). Stomatal density on the abaxial leaf surface was not affected by treatment in Arabis irsuta, Plantago lanceolata and Scabiosa columbaria, while on the adaxial surface it was significantly higher in plants grown under elevated CO2. The stomatal index was only calculated for Hippocrepis comosa and Plantago lanceolata, but in any case responded consistently to high CO2 (Table 2). In the mini-FACE system, both leaf surfaces of Cynodon dactylon, Avena barbata, Potentilla reptans and Convolvolus arvensis showed no significant differences in stomatal density between treatments (Table 1). In Medicago arabica and Trifolium repens stomatal density was significantly higher in control plants, but only for the adaxial leaf surface. The stomatal index was never influenced consistently by the treatment regardless of species (Table 2). At the CO2 spring in Strmec, averaged maximum (midmorning) leaf conductance (measured on different dates in April 1998 and September 1998 and 1999) showed consistently lower values in plants of Trifolium pratense, Polygonum hydropiper, Rumex crispus and Plantago lanceolata grown close to the CO2 spring, but insignificantly changed values in Tanacetum vulgaris and Echinochloa crus-galli (Fig. 3). At the Bossoleto experimental set-up, averaged maximum leaf conductance (measured during June 1996) was always negatively influenced by elevated CO2 in Scabiosa columbaria, Silene vulgaris and Plantago lanceolata (Fig. 3). In the mini-FACE system averaged maximum leaf conductance (measured in June, July and August 1999) was lower in plants of Medicago arabica and

187

Avena barbata grown in CO2-enriched rings. Cynodon dactylon displayed the opposite behaviour, while Trifolium repens and Convolvolus arvensis showed insignificant differences between treatments (Fig. 3). Medicago arabica was present in both CO2-enriched and control rings throughout the summer. A consistent trend in daily averaged maximum leaf conductance was followed throughout the season, always showing lower values in CO2-enriched than control rings but with differences between treatments decreasing as vapour pressure difference increased (Fig. 4). Diurnal trends in leaf water potential invariably showed a recovery during the night before predawn in both Cynodon dactylon and Trifolium repens, but did not show differences between control and CO2-enriched rings. An exception to this was lower predawn water potential in plants of Trifolium repens grown in control rings in late July (Fig. 5). Diurnal courses of leaf conductance resulted in clear differences between treatments in Trifolium repens, with consistently higher values in plants of the control than CO2-enriched rings throughout the day, while these differences were less marked for Cynodon dactylon and Convolvolus arvensis (Fig. 6). Increased CO2 concentration had inconsistent effects on leaf water potential components measured in midsummer on Cynodon dactylon and Medicago arabica (Table 3), except for tissue elasticity that was reduced and relative water content that was enhanced in plants of the former species grown in CO2-enriched rings (significantly in 1999). Total above-ground dry mass and cumulative leaf area index did not differ between treatments in all months and plant groups (Fig. 7). Root biomass and root density generally increased because of elevated CO2 exposure, though this was more evident in October than in June, and statistically significant only for the 15–30 cm soil layer (Fig. 8).

Table 2. Stomatal index in leaves (either on adaxial or abaxial surface) sampled at the CO 2 spring of Strmec (September–August 1999), at the Bossoleto experimental set-up (end of June 1996) and in the mini-FACE system (summer 1999, month depending on the species) Data are the means ± s.e. (units are in %). Percentage change between CO2 treatments [(elevated CO2 – ambient CO2)/ambient CO2] and P-level are also reported (CO2 concentrations are reported in the text) Abaxial elevated CO2 ambient CO2 Strmec Rumex crispus Plantago lanceolata Polygonum hydropiper Trifolium pratense Bossoleto Hippocrepis comosa Plantago lanceolata Mini-FACE Convolvolus arvensis Trifolium repens Medicago arabica

∆ (P-level) %

Adaxial elevated CO2 ambient CO2

∆ (P-level) %

12.45 ± 0.41 27.52 ± 0.86 18.20 ± 0.28 24.08 ± 0.81

13.00 ± 0.61 21.10 ± 0.88 17.62 ± 0.29 23.78 ± 0.75

–4.2 (ns) 30.4 (0.0000) 3.3 (ns) 1.3 (ns)

11.26 ± 0.49 26.18 ± 0.55 4.22 ± 0.43 18.31 ± 0.53

11.8 ± 0.49 26.7 ± 0.53 5.52 ± 0.42 16.3 ± 0.64

–4.6 (ns) –1.9 (ns) –23.6 (0.039) 12.3 (0.022)

24.31 ± 0.59 30.33 ± 1.72

23.86 ± 0.46 30.43 ± 0.94

1.9 (ns) –0.3 (ns)

23.96 ± 0.52 31.28 ± 0.54

23.72 ± 0.45 29.93 ± 0.65

1.0 (ns) 04.5 (ns)

19.36 ± 0.65 17.00 ± 0.77 19.10 ± 0.88

19.38 ± 0.67 18.00 ± 0.75 19.87 ± 0.98

–0.1 (ns) –5.6 (ns) –3.9 (ns)

20.00 ± 0.61 23.00 ± 0.38 23.54 ± 0.56

17.00 ± 0.66 22.48 ± 0.38 22.46 ± 0.55

17.6 (ns) 2.3 (ns) 4.8 (ns)


S. Marchi et al.

Discussion

Averaged maximum leaf conductance (mol m–2 s–1)

This study confirms that leaf conductance generally decreases under elevated CO2 concentrations, and that the effect is not transitory but persists over the long-term and Strmec

1.0 700 µmol mol

–1

0.8 ***

0.6

*** **

0.4

0.2

0.0

tu ce na ris Taulga v


***

360 µmol mol–1

0.4

a m hlo liu oc ifo se Tr aten chin galli pr E u scr

m

go um nta lata on r Pla ceo olyg pipe n o P r la d hy

x me s Ru ispu cr

Bossoleto 620 µmol mol 360 µmol mol–1 *

–1

***

0.3

0.2

***

0.1

0.0


Scabiosa columbaria

0.4

Plantago lanceolata

Silene vulgaris

through plant generations. The only exception is Cynodon dactylon, which showed higher leaf conductance values under elevated CO2 with respect to ambient conditions. Cynodon dactylon is a C4 species belonging to the Poaceae, and particularly common in Mediterranean areas where it is considered an important weed. The specific morphological habit of this species makes it particularly resistant to water stress; in fact, it was the most diffuse species in mini-FACE rings during the summer. C4 plants usually have higher water-use efficiencies than C3 species because of intrinsically lower leaf conductance (Pearcy and Ehleringer 1984), enabling them to adapt arid environments. An increase in leaf conductance in C4 species exposed to elevated CO2 has been observed for Andropogon gerardii Vitman (Le Cain and Morgan 1998). During periods of mild water stress, leaf conductance was higher in C3 control plants than in those exposed to elevated CO2, but treatment differences tended to decrease with the progress of seasonal and diurnal drought stress. Bunce (1998) suggested that, because abscisic acid sensitises stomata to CO2 concentration, abscisic acid might be involved in the response of stomata conductance to changes in leaf-to-air vapour pressure deficit. Osmotic potentials in Cynodon dactylon and Medicago arabica were not affected by elevated CO2, as observed in other grassland communities (Clark et al. 1999), indicating a lack of consistent osmotic adjustment for plants of CO2-enriched rings. A decrease in tissue elasticity for Cynodon dactylon grown in CO2-enriched rings, increasing relative water content, may be beneficial to maintain water uptake if a deeper root system promotes a constant water supply from soil during summer drought. A general reduction in leaf conductance under elevated CO2 concentrations could have important consequences for the ability of plants to endure water deficits as the restriction

Mini-FACE ***

580 µmol mol–1 360 µmol mol–1

0.3 *

**

0.2

0.1

0.0 Medicago arabica

Avena Convolvolus Trifolium barbata arvensis repens

Cynodon dactylon

Fig. 3. Averaged maximum leaf conductance (mid-morning) measured at the CO2 spring of Strmec (April 1998, and September 1998 and 1999), at the Bossoleto experimental set-up (June 1996) and in the mini-FACE system (June, July and August 1999); species are indicated in the text. Data are the means ± s.e. Significance is also reported: *, P < 0.05; **, P < 0.01; ***, P < 0.001; non-significant where unmarked. Growth CO2 concentration is indicated by symbols in the legend.

Daily averaged maximum leaf conductance (mol m–2 s–1)

188

Mini-FACE 0.6 0.5 0.4 0.3 0.2 0.1 0.0 140

580 µmol mol–1 360 µmol mol–1 150

160

170

180

190

200

210

220

Day of year

Fig. 4. Daily averaged maximum leaf conductance for Medicago arabica (mid-morning) measured in the mini-FACE rings during summer 1999. Data are the means ± s.e. Differences between CO2 treatments were always significant. Growth CO2 concentration is indicated by symbols in the legend.



Cynodon dactylon

Trifolium repens

0

0

20 June 2000 Leaf water potential (MPa)

Leaf water potential (MPa)

20 June 2000

–1

–2

–1

–2

580 µmol mol–1

580 µmol mol–1

360 µmol mol–1

360 µmol mol–1

–3

–3 2

4

6

8

10

12

14

16

18

2

4

6

Time of day (hour)

8

10

12

0

18

7 July 2000 Leaf water potential (MPa)

Leaf water potential (MPa)

16

0

7 July 2000

–1

–2

–3

–1

–2

–3 2

4

6

8

10

12

14

16

18

2

4

6

Time of day (hour)

8

10

12

14

16

18

Time of day (hour)

0

0



14

Time of day (hour)

–1

–2

–3

–1

–2

–3 2

4

6

8

10

12

Time of day (hour)

14

16

18

2

4

6

8

10

12

14

16

18

Time of day (hour)

Fig. 5. Diurnal trends in leaf water potential in Cynodon dactylon and Trifolium repens grown in mini-FACE, control and CO2-enriched rings, during July 2000. Data are the means ± s.e. Growth CO2 concentration is indicated by symbols in the legend.

189

S. Marchi et al.

Above ground biomass (g m–2)


0.5

0.4

0.3

0.2


600 500 400 300 200 100 0

Poaceae Fabaceae Others

8

10

12

14

16

18

Time of day (hour) 0.5

Cynodon dactylon 0.4

0.3


10 June 2000

0.0

Leaf conductance (mol m–2 s–1)

May

700

Total 580 µmol mol–1 360 µmol mol–1

500

August 400 300 200 100 0

Total


0.2

0.1

10 July 2000 0.0 8

10

12

14

16

18

Time of day (hour) 0.5

Trifolium repens Leaf conductance (mol m–2 s–1)

800

0.1


Leaf conductance (mol m–2 s–1)

Convolvolus arvensis

300

October

250 200 150 100 50 0

0.4

Total


0.3

0.2

0.1

10 July 2000 0.0 8

10

12

14

16

18

Time of day (hour)

Fig. 6. Diurnal trends in leaf conductance in Cynodon dactylon, Trifolium repens and Convolvolus arvensis grown in mini-FACE, control and CO2-enriched rings, during July 2000. Data are the means ± s.e. Growth CO2 concentration is indicated by symbols in the legend.


190

7 6 5 4 3 2 1 0

May

August

October

Fig. 7. Total above-ground dry mass after harvest in May, August and October, and cumulative leaf area index measured in the mini-FACE, control and CO2-enriched rings, in 2000. Data are the means ± s.e. Growth CO2 concentration is indicated by symbols in the legend.



of water loss is likely to benefit turgor maintenance and limit dehydration damages (Morison 1998). However, leaf level measurements (though on species growing in close association) should be considered with caution when extending decreased stomatal conductance to canopy-scale transpiration because of numerous atmospheric feedbacks and characteristic variations of particular functional groups in sensitivity to increased CO2. The proportional effect of CO2 enrichment on stomatal conductance might become negligible as stomata close in response to environmental stresses. Indeed, the response of plants to summer drought in high CO2 depends on the balance of the CO2-induced decrease in stomatal conductance and the increase in water use caused by augmented leaf area. Whether or not a crop canopy actually uses more or less soil water when grown in elevated CO2 depends on stomatal conductance, leaf area production, and various environmental factors that affect the energy balance. In this study, cumulative leaf area index (thus total transpiring surface area) and total above-ground biomass did not differ between treatments. Plant growth is altered by changes in water potential (Yegappan et al. 1982), and a positive relationship between leaf growth and turgor may be expected after elevated CO2 exposure due to a likely increase in water use efficiency. Nevertheless, treatmentinduced variations in leaf water potential were negligible. In wet soil, a reduction in leaf conductance under high CO2 without an increase in leaf water potential implies a decrease in hydraulic conductance. A decline in plant hydraulic conductance might have a structural basis (e.g. through anatomical changes), but it could also represent a homeostatic response to lower transpiration rates under elevated CO2 during growth (Bunce and Ziska 1998). Resistance to drought may depend upon factors affecting the balance of evaporative demand and supply, including foliage area and xylem water transport capacity. Biomass and root density in these grasslands increased because of elevated CO2 expo-

191

sure, which could facilitate exploration of a greater soil volume and constitute a sink for extra available carbon (Newton et al. 1996). Deeper roots, on the other hand, might enhance plant water consumption, thereby increasing the potential for exposure to drought (Eamus 1996). Long-term components may add to short-term ones, potentially contributing to any measured difference in stomatal conductance when plants are grown in elevated CO2 concentration. Changes in leaf conductance can be caused by variations in number or frequency of stomata as individual leaves expand, and at the whole-plant scale by changes in total leaf area (leaf size or numbers) as the plant develops (Morison 1998). The decrease of stomatal conductance under elevated CO2 of the studied plants, however, was not associated with changes in stomatal number and distribution on leaf surfaces. Experiments under controlled conditions (Woodward 1987), herbarium specimens (Peñuelas and Matamala 1990), fossil leaves (Beerling and Woodward 1993) and leaves preserved in pack rat middens (Van de Water et al. 1994) have indicated a general reduction in stomatal density as CO2 level increased since the last glacial maximum. Other authors have observed no consistent effect of high CO2 on stomatal density (Estiarte et al. 1994). Conclusions from stomatal counts on fossil and herbarium specimens must be treated with extreme caution, since these studies rarely take into account natural variations and technical artefacts (Poole et al. 1996). The absence of consistent responses in stomatal density to changing CO2 concentration in manipulative experiments might depend on the short-term exposure (Malone et al. 1993). In addition, experiments in monoculture may be misleading, since plants do not experience competition (Bazzaz 1990). Our study suggests only minor responses or no significant changes in stomatal number even after long-term exposure of these grassland communities to a CO2 level exceeding current ambient levels. Bettarini et al. (1998) found that stomatal

Table 3. Tissue water relations of Cynodon dactylon and Medicago arabica sampled in the mini-FACE system (July 1999 and 2000) Data are the means ± s.e. (n = 8–20). Percentage change between CO2 treatments [(elevated CO2 – ambient CO2)/ambient CO2] and P-level are also reported (CO2 concentrations are reported in the text) Cynodon dactylon elevated CO2 ambient CO2 1999 R*tlp (%) πsat (MPa) πtlp (MPa) Rs (%) ε (MPa) 2000 R*tlp (%) πsat (MPa) πtlp (MPa) Rs (%) ε (MPa)

∆ (P-level) %

95.8 ± 0.9 –1.35 ± 0.06 –1.46 ± 0.05 60.5 ± 7.0 42.3 ± 7.5

91.8 ± 1.6 –1.27 ± 0.04 –1.52 ± 0.06 51.1 ± 3.6 19.1 ± 2.4

4.3 (0.052) 6.6 (ns) –3.9 (ns) 18.3 (ns) 121.2 (0.008)

90.0 ± 1.9 –1.56 ± 0.03 –1.83 ± 0.07 73.8 ± 3.2 33.1 ± 8.0

88.7 ± 2.2 –1.53 ± 0.05 –1.79 ± 0.08 83.7 ± 5.0 25.4 ± 4.7

1.4 (ns) 1.9 (ns) 2.1 (ns) –11.8 (ns) 30.3 (ns)

Medicago arabica elevated CO2 ambient CO2 80.6 ± 2.5 –1.14 ± 0.06 –1.52 ± 0.04 79.6 ± 2.8 7.4 ± 1.1

80.1 ± 1.7 –1.11 ± 0.04 –1.54 ± 0.04 71.1 ± 3.3 6.4 ± 0.75

∆ (P-level) % 0.7 (ns) 3.0 (ns) –1.3 (ns) 11.9 (ns) 16.0 (ns)

192


S. Marchi et al.

density was unaffected in most of the 17 examined species growing in a CO2 spring, although stomatal conductance was generally lower under elevated CO2. In the mini-FACE experiment, the only variation in stomatal density was for the adaxial leaf surface of Medicago arabica and Trifolium repens, but the stomatal index never differed between treatments. Again, variations in stomatal density between treatments for the adaxial leaf surface of Arabis hirsuta, Plantago lanceolata and Scabiosa columbaria did not correspond to changes in the stomatal index. Discrepancies between stomatal density and index might depend on increased epidermal cell expansion leading to a larger final cell size due to ameliorated plant water status after high CO2 exposure (lower stomatal conductance and higher turgor pressure). Variations in stomatal index may be expected only after changes in differentiation of guard cells. Plants grown in short-term experiments were not exposed for generations to elevated CO2 concentration, as in the case of natural CO2 springs, and stomatal response might be limited by the degree of associated phenotypic plasticity. Nevertheless, the plasticity in morphological and physiological characters in grassland communities should be high owing to the extreme inter-annual climatic varia-

Root biomass (g m–2)

400


350 300 250 200 150

*

100

*

50 0 0–15 cm

15–30 cm

June

0–15 cm

15–30 cm

October

Root density (m m–2)

2500

2000

1500

* *

1000

500

0 0–15 cm

June

15–30 cm

0–15 cm

15–30 cm

October

Fig. 8. Root biomass and root density (0–15 and 15–30 cm soil layers) measured in the mini-FACE, control and CO2-enriched rings, after harvest in June and October 2000. Data are the means ± s.e. Growth CO2 concentration is indicated by symbols in the legend.

bility of these systems (Knapp et al. 1994). The increase in atmospheric CO2 concentration did not cause adaptive variations either in stomatal number or in cell division of these grassland communities, in which other factors might be more limiting than CO2 (Körner 1993). Changes in stomatal conductance after high CO2 exposure are probably due to modifications in stomatal aperture independently of stomatal frequency on the leaf surface (Murray 1995). Indeed, ontogenetic development of plants begins at meristem level with cell division, distension and differentiation, which are independent processes determined genetically and environmentally (Taylor 1997). In conclusion, the plants studied have an intrinsic capacity for sagacious water use compared with carbon economy. Evolution adapted the form and function of plants, making all resources equally limiting for growth and development (Bloom et al. 1985). An increase of atmospheric CO2 concentration results in relative changes of carbon availability, water conservation becoming somewhat more important in order to endure severe periodic drought at Mediterranean latitudes. In drought conditions, under elevated CO2 stomatal regulation might act as an environmental buffer for modulating evapotranspiration, only partially compensated for by changes in leaf area (Idso and Idso 1994). In regions with concurrently limiting soil water and nutrient resources, and where competition for light already limits growth, the response of grassland communities to CO2 enrichment in terms of leaf area index will probably be negligible, as in the present case, and should be interpreted within an ecological context (Diaz 1995). Any reduction in leaf conductance would decrease transpiration per unit of leaf area. Under elevated CO2, limited leaf cooling might enhance canopy temperature and affect water consumption in C3 species differently from C4 species, varying community productivity (Epstein et al. 1997). In the long term, changes in competitive ability might result in variations of distribution and relative abundance of these species, altering community composition (Teugels et al. 1995). Offsetting effects and species-specific responses to a rise in atmospheric CO2 level make predictions of elevated CO2-induced consequences on stomatal characters and water use difficult. It can be hypothesised, however, that the proportional effect of CO2 enrichment on stomatal conductance of most species in these grassland communities becomes negligible as stomata close in response to environmental stresses. Co-occurring plants may have compensatory regulations of gas exchange and water relations, minimising interspecific variation on a larger scale. Acknowledgments We thank Anna Longobucco and Silvia Sforzi for technical assistance in field measurements and site maintenance of mini-FACE systems. The work was supported by the E.U. ‘Megarich’ project.


References Amthor JS (1995) Terrestrial higher-plant response to increasing atmospheric [CO2] in relation to the global carbon cycle. Global Change Biology 1, 243–274. Bazzaz FA (1990) The response of natural ecosystems to the rising global CO2 level. Annual Review of Ecology and Systematics 21, 167–196. doi:10.1146/ANNUREV.ES.21.110190.001123 Beemster GTS, Baskin TI (1998) Analysis of cell division and elongation underlying the developmental acceleration of root growth in Arabidopsis thaliana. Plant Physiology 116, 1515–1526. doi:10.1104/PP.116.4.1515 Beerling DJ, Woodward FI (1993) Ecophysiological responses of plants to global environmental change since the last glacial maximum. New Phytologist 125, 641–648. Bettarini I, Vaccari FP, Miglietta F (1998) Elevated CO2 concentrations and stomatal density: observations from 17 plant species growing in a CO2 spring in central Italy. Global Change Biology 4, 17–22. doi:10.1046/J.1365-2486.1998.00098.X Bloom AJ, Chapin FS, Mooney HA (1985) Resource limitation in plants — an economic analogy. Annual Review of Ecology and Systematics 16, 363–392. Bunce JA (1998) Effects of environment during growth on the sensitivity of leaf conductance to changes in humidity. Global Change Biology 4, 269–274. doi:10.1046/J.1365-2486.1998. 00148.X Bunce JA, Ziska LH (1998) Decreased hydraulic conductance in plants at elevated carbon dioxide. Plant, Cell and Environment 21, 121–126. doi:10.1046/J.1365-3040.1998.00256.X Clark H, Newton PCD, Barker DJ (1999) Physiological and morphological responses to elevated CO2 and a soil moisture deficit of temperate pasture species growing in an established plant community. Journal of Experimental Botany 50, 233–242. doi:10.1093/JEXBOT/50.331.233 Cook AC, Tissue DT, Roberts SW, Oechel WC (1998) Effects of longterm elevated [CO2] from natural CO2 springs on Nardus stricta: photosynthesis, biochemistry, growth and phenology. Plant, Cell and Environment 21, 417–425. doi:10.1046/J.1365-3040.1998. 00285.X Curtis PS, Drake BG, Leadley PW, Arp WJ, Whigham DF (1989) Growth and senescence in plant communities exposed to elevated CO2 concentrations on an estuarine marsh. Oecologia 78, 20–26. Diaz S (1995) Elevated CO2 responsiveness, interactions at the community level and plant functional types. Journal of Biogeography 22, 289–295. Drake BG, Gonzàlez-Meler MA, Long SP (1997) More efficient plants: a consequence of rising atmospheric CO 2? Annual Review of Plant Physiology and Plant Molecular Biology 48, 609–637. doi:10.1146/ANNUREV.ARPLANT.48.1.609 Eamus D (1996) Responses of field grown trees to CO2 enrichment. Commonwealth Forestry Reviews 75, 39–47. Epstein HE, Lauenroth WK, Burke IC, Coffin DP (1997) Productivity patterns of C3 and C4 functional types in the US Great Plains. Ecology 78, 722–731. Estiarte M, Peñuelas J, Kimbal BA, Idso SB, LaMorte RL, Pinter PJ Jr, Wall GM, Garcia RL (1994) Elevated CO2 effects on stomatal density of wheat and sour orange trees. Journal of Experimental Botany 45, 1645–1668. Ferris R, Taylor G (1994) Elevated CO2, water relations and biophysics of leaf extension in four chalk grassland herbs. New Phytologist 127, 297–307. Field CB, Jackson RB, Mooney HA (1995) Stomatal response to increased CO2: implications from the plant to the global scale. Plant, Cell and Environment 18, 1214–1225.


193

Hendrey GR, Lewin KF, Nagy J (1993) Free Air Carbon Dioxide Enrichment: development, progress, results. Vegetatio 104/105, 17–31. Hovenden MJ (2003) Photosynthesis of coppicing poplar clones in a free-air CO2 enrichment (FACE) experiment in a short-rotation forest. Functional Plant Biology 30, 391–400. doi:10.1071/ FP02233 Idso KE, Idso SB (1994) Plant responses to atmospheric CO 2 concentration enrichment in the face of environmental constraints: a review of the past 10 years’ research. Agricultural and Forest Meteorology 69, 153–202. doi:10.1016/0168-1923(94)90025-6 Jackson RB, Sala OE, Field CB, Mooney HA (1994) CO2 alters water use carbon gain, and yield in a natural grassland. Oecologia 98, 257–262. Keeling CD, Whorf TP, Whalen M, van der Plicht J (1995) Interannual extremes in the rate of rise of atmospheric carbon dioxide since 1980. Nature 375, 666–670. doi:10.1038/375666A0 Knapp AK, Cocke M, Hamerlynck EP, Owensby CE (1994) Effect of elevated CO2 on stomatal density and distribution in a C4 grass and a C3 forb under field conditions. Annals of Botany 74, 595–599. doi:10.1006/ANBO.1994.1159 Körner C (1988) Does global increases of CO2 alter stomatal density? Flora 181, 253–257. Körner C (1993) Scaling from species to vegetation: the usefulness of functional groups. In ‘Biodiversity and ecosystem function’. (Eds HA Mooney and E-D Schulze) pp. 117–140. (Springer-Verlag: Berlin) Körner C, Miglietta F (1994) Long term effects of naturally elevated CO2 on Mediterranean grassland and forest trees. Oecologia 99, 343–351. Le Cain DR, Morgan JA (1998) Growth, gas exchange, leaf nitrogen and carbohydrate concentrations in NAD-ME and NADP-ME C4 grasses grown in elevated CO2. Physiologia Plantarum 102, 297–306. doi:10.1034/J.1399-3054.1998.1020218.X Malone SR, Mayeux HS, Johnson HB, Polley HW (1993) Stomatal density and aperture length in four plant species grown across a subambient CO2 gradient. American Journal of Botany 80, 1413–1418. Maek I, Pfanz H, Bati F, Vodnik D (2003) Growth and root respiration of C4 plants under CO2 enrichment. Acta Biologica Slovenica (in press). Miglietta F, Raschi A, Bettarini I, Resti R, Selvi F (1993) Natural CO2 springs in Italy: a resource for examining long-term response of vegetation to rising atmospheric CO2 concentrations. Plant, Cell and Environment 16, 873–878. Miglietta F, Lanini M, Bindi M, Magliulo V (1997) Free air CO2 enrichment of potato (Solanum tuberosum L.): design and performance of the CO2-fumigation system. Global Change Biology 3, 417–427. doi:10.1046/J.1365-2486.1997.00076.X Miglietta F, Hoosbeck MR, Foot J, Gigon F, Hassinen A, Heijmans M, Peressotti A, Saarinen T, van Bremen N, Wallén B (2001) Spatial and temporal performance of the mini-FACE (free air CO2 enrichment) system on bog ecosystems in Northern and Central Europe. Environmental Monitoring and Assessment 66, 107–127. doi:10.1023/A:1026495830251 Morison JIL (1998) Stomatal response to increased CO2 concentration. Journal of Experimental Botany 49, 443–452. doi:10.1093/ JEXBOT/49.SUPPL_1.443 Murray DR (1995) Plant response to carbon dioxide. American Journal of Botany 82, 690–697. Newton PCD, Bell CC, Clark H (1996) Carbon dioxide emissions from mineral springs in Northland and the potential of these sites for studying the effects of elevated carbon dioxide on pastures. New Zealand Journal of Agricultural Research 39, 33–40.

194


S. Marchi et al.

Overdieck D, Reining F (1986) Effect of atmospheric CO2 enrichment on perennial ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.) competing in managed model-ecosystems. Acta Oecologica. Oecologia Plantarum 7, 357–366. Parton WJ, Scurlock JMO, Ojima DS, Gilmanov TG, Scholes RJ, et al. (1993) Observations and modeling of biomass and soil organic matter dynamics for the grassland biome worldwide. Global Biogeochemical Cycles 7, 785–809. Pearcy RW, Ehleringer JR (1984) Comparative ecophysiology of C3 and C4 plants. Plant, Cell and Environment 7, 1–13. Pearson M, Davies WJ, Mansfield TA (1995) Asymmetric responses of adaxial and abaxial stomata to elevated CO2: impacts on the control of gas exchange by leaves. Plant, Cell and Environment 18, 837–843. Peñuelas J, Matamala R (1990) Changes in N and S leaf content, stomatal density and specific leaf area of 14 plant species during the last three centuries of CO2 increase. Journal of Experimental Botany 41, 1119–1124. Poole M, Weyers JDB, Lawson T, Raven JA (1996) Variation in stomatal density and index: implications for paleoclimatic reconstruction. Plant, Cell and Environment 19, 705–712. Sanità di Toppi L, Marabottini R, Badiani M, Raschi A (2002) Antioxidant status in herbaceous plants growing under elevated CO2 in mini-FACE rings. Journal of Plant Physiology 159, 1005–1013. Schlesinger WH (1991) ‘Biogeochemistry: an analysis of global change’. (Academic Press: New York) Schulte PJ, Hinckley TM (1985) A comparison of pressure–volume curve data analysis techniques. Journal of Experimental Botany 36, 1590–1602. Seneweera SP, Ghannoum O, Conroy JP, Ishimaru K, Okada M, Lieffering M, Yong Kim H, Kobayashi K (2002) Changes in source–sink relations during development influence photosynthetic acclimation of rice to free air CO2 enrichment (FACE). Functional Plant Biology 29, 947–955. doi:10.1071/PP01250 Taylor CB (1997) Plant vegetative development: from seed and embryo to shoot and root. The Plant Cell 9, 981–988. doi:10.1105/TPC.9.7.981 Teugels H, Nijs I, Van Hecke P, Impens I (1995) Competition in a global change environment: the importance of different plant traits for competitive success. Journal of Biogeography 22, 297–305.

Tognetti R, Raschi A, Jones MB (2000) Seasonal patterns of tissue water relations in three Mediterranean shrubs co-occurring at a natural CO2 spring. Plant, Cell and Environment 23, 1341–1351. doi:10.1046/J.1365-3040.2000.00645.X Vaccari FP, Bettarini I, Giuntoli A, Miglietta F, Raschi A (2001) Mediterranean grassland community under elevated CO2: observations from a CO2 spring. Journal of Mediterranean Ecology 2, 41–50. Van de Water PK, Leavitt SW, Betancourt JL (1994) Trends in stomatal density and 13C/ 12C ratios of Pinus flexilis needles during the last glacial–interglacial cycle. Science 264, 239–243. Van Volkenburgh E (1999) Leaf expansion: an integrating plant behaviour. Plant, Cell and Environment 22, 1463–1473. doi:10.1046/J.1365-3040.1999.00514.X Vazzana C, Puliga S, Bochicchio A, Raschi A (1988) Studio della densità stomatica in un ecotipo di erba medica (Medicago sativa L.). Rivista di Agronomia 22, 127–132. Vodnik D, Pfanz H, Maček I, Kastelec D, Lojen S, Batič F (2002) Photosynthesis of cockspur [Echinochloa crus-galli (L.) Beauv.] at sites of naturally elevated CO2 concentration. Photosynthetica 40, 575–579. doi:10.1023/A:1024308204086 Ward JK, Tissue DT, Thomas RB, Strain BR (1999) Comparative responses of model C3 and C4 plants to drought in low and elevated CO2. Global Change Biology 5, 857–867. doi:10.1046/ J.1365-2486.1999.00270.X Woodward FI (1987) Stomatal numbers are sensitive to increasing in CO2 from preindustrial levels. Nature 327, 617–618. doi:10.1038/327617A0 Woodward FI, Kelly CK (1995) The influence of CO2 concentration on stomatal density. New Phytologist 131, 311–327. Woodward FI, Lake JA, Quick WP (2002) Stomatal development and CO2: ecological consequences. New Phytologist 153, 477–484. doi:10.1046/J.0028-646X.2001.00338.X Yegappan TM, Paton DM, Gates CT (1982) Water stress in sunflower (Helianthus annuus L.) 2. Effects on leaf cells and leaf area. Annals of Botany 49, 63–68.

Manuscript received 28 July 2003, received in revised form 24 October 2003, accepted 4 December 2003

http://www.publish.csiro.au/journals/fpb